So with sipp 3.995 (the last version available at sourceforge, which is
closest to the 3.3 I was using before migrating to the new laptop), the
|ontimeout| attribute works as expected (i.e. when the timeout elapses,
the scenario execution jumps to the label indicated in the |ontimeout|
attribute - see the scenario screen below), but as I've suggested
earlier, the presence of |<recv response´"xxx" optional="true"/>| before
the |<recv response="200" timeout="3000" ontimeout="some_label"/>|
causes the timeout not to ever happen. Which I've concluded to be quite
logical back then when I was dealing with it myself, and as I could work
that around, I gave up trying to understand that in detail quite quickly.
Now, I gave it a bit more, so I've configured also the 100 and 180 with
|timeout| and |ontimeout|, where the timeout for the 100 was different
from the one for the 180 and the 200 (I've commented out the rest of
optional responses). Regardless whether the 100's timeout was longer or
shorter than the one for the 180 and the 200, it always fired first.
So there are two differences in 3.6 as compared to 3.3:
* optional messages preceding the mandatory one with ontimeout now
don't prevent the timer from firing (improvement),
* the ontimeout value is not used (regression).
Given that, I'd think that someone has made modifications to the
handling of optional messages when a timeout is specified for the
mandatory one following them, and this may have included some mistake in
processing of the ontimeout attribute.
Hence I suggest you to raise an issue at github. Maybe with a note that
it should be selectible whether a expiration of this particular timeout
is considered a call failure or not, as when we want to send a BYE only
if the remote peer didn't within some time, it may not always be
considered a failure.
Other than that, sipp kept saying it cannot load or parse your scenario
file until I've copy-pasted it to another one using a text editor.
Changing iso-8859-2 to iso-8859-1 in the original file did not help.
|<label id="3"/>| was missing in it, but with the copy, sipp just
complained about absence of label 3, not about a problem with the whole
file. Weird. So maybe do the same and try again before filing a bug, the
3.6 parser has been reworked as compared to 3.3, so the same problem
with the file may cause a different outcome there.
P.
|------------------------------ Scenario Screen -------- [1-9]: Change
Screen --||
|| Call-rate(length) Port Total-time Total-calls Remote-host||
|| 10.0(0 ms)/1.000s 5060 15.20 s 1 127.0.0.1:5082(UDP)||
||
|| Call limit reached (-m 1), 1.006 s period 16 ms scheduler resolution||
|| 1 calls (limit 615) Peak was 1 calls, after 0 s||
|| 0 Running, 2 Paused, 4 Woken up||
|| 0 dead call msg (discarded) 0 out-of-call msg (discarded)||
|| 3 open sockets||
|| 0 Total RTP pckts sent 0.000 last period RTP rate
(kB/s)||
||
|| Messages Retrans Timeout
Unexpected-Msg||
|| REGISTER ----------> 1 0||
|| 200 <---------- 1 0 0 0||
|| 500 <---------- 0 0 0 0||
|| 401 <---------- 0 0 0 0||
|| REGISTER ----------> 0 0||
|| 200 <---------- 0 0 0 0||
|| Pause [ 4000ms] 1 0||
|| INVITE ----------> 1 0||
|| 200 <---------- 0 0 1 0||
||
|| ACK ----------> 0 0||
|| Pause [ 500ms] 0 0||
|| [ NOP ]||
|| Pause [ 8000ms] 0 0||
|| BYE ----------> 1 0||
|| 200 <---------- 0 0 0 0||
||
|| Pause [ 4000ms] 0 0||
|| REGISTER ----------> 0 0||
|| 200 <---------- 0 0 0 0||
|| 500 <---------- 0 0 0 0||
|| 480 <---------- 0 0 0 0||
|| 401 <---------- 0 0 0 0||
|| REGISTER ----------> 0 0 0||
|| 200 <---------- 0 0 0 0||
|| Pause [ 4000ms] 0 0||
||------- Waiting for active calls to end. Press [q] again to force
exit. -------||
|
_______________________________________________
Sipp-users mailing list
Sipp-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sipp-users
